The method and apparatus of controlling air-fluid distribution and heat exchange may apply to any commercial, industrial, scientific, or engineering application wherein air flow, fluid flow, gas flow, containment or mixture thereof would require most efficient, most precise distribution, articulation, and delivery. However, the main application as described herein will namely address the HVAC (Heating, Ventilating, Air Conditioning) industry.
The following description and claims are supported by established facts known from scientific and engineering principles as set forth by the laws of fluid dynamics, fluid statics, thermal dynamics, affinity laws, and by building and energy codes.
The Primary Mover
The first step in the process of determining system status begins with the primary mover and air handler (or fluid handler) itself, including all of its internal components. Referring to FIG. 2, 2A, 2B, these illustrations depict an “old school” arrangement of mover testing for TP, SP, and Vp (Total Pressure, Static Pressure, and Velocity Pressure [of mover.]) It will establish a premise of known methodology, which will be referred to throughout the specification.
The various testing elements (probes) are arranged at the center of each duct. Note that there is no indication of whether these are meant to suggest a traverse of each duct or a testing at their cross-sectional center points (V-max or maximum velocity.) This also becomes moot when viewing FIG. 2A, as a true static pressure acts laterally against the walls of a duct, not over its cross-section, though some negligible force may be sensed there with a static probe. It would then, therefore, be logical to state that where the velocity is maximal, the static pressure would be minimal. The other assumption in this sensing arrangement is that the cross sections of discharge and suction have laminar flow, which in the case of most centrifugal fans, it certainly would not, particularly on its inlet side in close proximity to the fan. This is why sensors and flow stations must be located a sufficient distance downstream or upstream of the mover and with adequate straight section of duct or piping run.
Ready comparisons may be drawn between these early figures and FIG. 13, 14, 14A, 14B, primary mover sensor logic as employed by the described method and apparatus, which takes these fundamentals further and broadens their scope. These are schematic depictions of the sensor arrangements whose actual configuration may differ in appearance, though the principle function remains. Various sensor stations, assemblies, and “grids,” as we will call them, currently exist that may appear vastly different from either an equal area or log traverse, though the comprising elements (static, impact sensors) must be the same or they must be incorrect, though they may be somewhat functional with corrective calibration. References are made according to known and accepted methods of testing.
Referring also to FIG. 15, 15A, 15B, terminal or in-line device sensor logic, one key difference between a mover and its terminal device when making a dynamic (Vp) comparison under lab conditions with no system attached, is that the mover's flow-volume can only be measured on one side. Being an active device and a constant volume machine, its manometer reading (or differential) would otherwise equal neutral or zero.
A static differential comparison where a constant volume mover is concerned will be contingent, as this will be largely dependent on whether the inlet remains open to atmosphere (entirely in the form of velocity and, thus, negated) or ducted to some degree. Additionally, the percent “wide open” testing will have an impact on this arrangement. As different degrees (or percentages) of closure are applied to the mover, the static content will shift more from one side to another under varying conditions. Its total amount will remain potentially, but conversion and shifting will occur. And, this will affect namely how much “system” may be applied to the suction of the mover, where system design length of run per cross-section is concerned. The optional sensor arrangements shown have to do with already packaged or housed existing systems that may incur SP or Vp losses on one or the other side of the mover.
Undoubtedly, the type of mover will have an impact on test methods. For example, an axial fan or positive displacement pump will lean towards pressure constancy inlet to outlet, while centrifugal movers will exhibit more flexibility because of the nature of their construction and the forces at work. Mover aside, the described methodology clearly holds for the terminal device, particularly through its range of motion and with the mover's total power applied as a constant or variable.
One key difference in the diagram shown in FIGS. 2, 2A, and 2B, is that the SP and Vp readings in determining “Fan SP” and “Fan Vp” seem to be slanted toward only the discharge of the mover, in so far as each is concerned. This probably assumes inlet open to atmosphere (100% dynamic flow) on the mover's suction side with little or no ducting, ideally suited to an open plenum return, perhaps. Lab testing standards typically use this condition: open inlet with ducted discharge.
In the case of FIG. 2, it is safe to assume that the dynamic aspect is negated by the total impact sensing on the inlet, though this negates SP on this side as well, especially once ducted and how ducted. Typically speaking, however, when one side of a mover is 0.00″ WC static (or 100% velocity,) the other side is deemed to be 100% of its static power. But analyzing these effects are crucial to avoiding the pitfalls of presumption.
Additionally, the arrangement doesn't account for 1) System Effect losses once the mover is fitted and packaged. 2) The characteristic ductwork, namely on the suction side and the effect it will have on the mover, totally speaking. 3) There is no apparent reference to atmosphere wherein TP and SP are concerned, and establishing this may be difficult considering that the interior of building envelopes will taint the results, for the very reasons described in this specification.
The aim here, however, is not to play out differences, but rather describe how the said method and apparatus refers to known principles and progresses from these as a valid starting point to those already schooled in “the art” and provide a logical background to its development for clearer understanding.
The Fan Total Pressure
The Fan Total Pressure is a core measurement of the primary mover's total strength or total muscle, internally speaking. This determination is crucial to sizing the air-fluid distribution system in its entirety, full circle—discharge to suction—and, subsequently, establishing the representative system curve connected to the primary mover. This reading is taken directly at the mover's inlet and outlet with no other elements between. FIG. 3 shows a schematic of a typical “draw-through” unit with this demarcation and others delineated across its profile.
As shown in this example of a typically packaged or housed system, each component has a section. Firstly, we find the mixing box, where return air and outdoor air enter and mix airstreams; or simply return air alone, whether in the form of 100% return air or containing some percentage of outdoor air content. It may also contain an added air stream or fluid content supplied (ducted in) at some point upstream. The next section, moving in the direction of suction flow, is typically a filter or pre-filter section, followed by the cooling or heating coil itself, where primary heat exchange takes place. Following these, the blower cabinet and, finally, discharge. In some cases, there may be additional segments aft of the blower (filters, additional coils, etc.) It is here, however, exactly at the primary mover's inlet, where one sensor grid is connected and the other at the fan's discharge in determining a Fan Total Pressure.
In the past, with “built up” systems, i.e. systems that didn't arrive from the manufacturer with cabinets and housings, but were rather just blowers, motors, drives, and other basic components for field assembly, the traditional method of determining Total Fan Power was to arrange an impact tube (total pressure sensing element) at both the fan's ducted inlet and its ducted discharge. For a proper “Fan Total Pressure” to be taken, these two impact tubes were connected directly to a manometer (HI+ and LO−) and, hence, the total “muscle” of the blower was determined by the manometer differential in “WC or “WG units (same denotation.) Similarly, a “Fan Static Pressure,” to use generic terms, would be determined by a static sensor at its outlet, minus total pressure (impact sensor) at its inlet as a differential across both manometer connections. Again, refer to FIGS. 2 and 2A.
However, with modern “packaged” systems, blower mounting and housing inside of a cabinet has made this process vary considerably. For practical purposes, the new meaning accepted or simply understood by manufacturers and design engineers is that the blower's “Total Pressure” is simply measured as two “added” static pressure readings directly at the blower inlet and its discharge, these actually being subtracted (differentiated) as a negative and positive; for example, +5 “WC read at outlet minus −5 “WC at suction inlet equaling 10 (5−−5, or 5+5, a double negative thus added.) This can also be thought of as two absolute values, since it represents the fan's total power, coming and going combined.
Though technically, this is not the tried and true method, since it only considers static forces and not dynamic ones, it is the widely used method and has been employed for practical field measurement purposes, so long as the manufacturer's, design engineer's, and balancing agency's understandings are the same, thus the idea is corroborated and the intentions are the same. The design engineer, manufacturer, and balancers, however, should be aware of this fact for serious consideration when selecting, supplying, and testing the equipment, respectively, so the dynamic aspect of this equation is not overlooked. This point is stressed by the known fact that field measured Static Pressure readings are considered among the least reliable data in an existing or “as-built” system.
Furthermore, the immediate discharge in close proximity to a blower is primarily in the form of pure, non-uniform velocity, until static regain occurs approximately ⅔ of the way into the system, when there is a system. This fact alone may contribute to misleading or misinterpreted test results as well. Though in terms of static measurement, a higher static reading will occur at the enclosed inlet to somewhat compensate for this, reflecting the fan's total static power if only on one side, and with the added proviso that those are the terms agreed upon.
The recommended standard for testing any type of fluid flow is a uniform, stable condition known as laminar flow, normally occurring 2.5 duct widths for every 2500 FPM or less of discharge velocity from a mover and 1 additional duct width for every additional 1000 FPM. It is also accepted that there should be no more than 15 degrees converging or 7 degrees diverging in any fittings under such conditions. This is an equivalent round duct diameter, whereby a rectangular fitting would be converted through: SQ. RT. 41 w/PI. This criterion is also known as the 100% effective duct length, through which it is supposed that the total effectiveness of the mover may be realized.
The traditional method (two impact tubes) may have been employed where such systems offered an inlet duct run directly into the blower inlet where possible. In-line axial and radial-type centrifugal fans, both being ducted in series, end to end, may have been tested this way, so long as differences were noted and understood when compared to dissimilar systems. Those skilled and experienced in the art, such as HVAC engineers or Testing & Balancing Supervisors should be aware of these differences.
It is understood, for example, that packaged units are assigned an ESP (External Static Pressure) and that simpler movers, such as fans with no filters, coils, or other sectional devices fore or aft of the mover itself are understood to be assigned with what is both an ESP and TSP (Total Static Pressure,) these becoming one and the same concept because of no internal component losses coming into play.
These concepts still remain the source of much debate in the industry, and as a result, no consistent air-fluid distribution control system has been adequately or consummately applied, but rather the emphasis has been more on temperature control alone. Aside from this fact alone, this is true for many more reasons, which will be discussed in various sections of the following specification.
Practically speaking, this outdated terminology will be cited more carefully since it produces a conflict in terms: Total Pressure, Total Fan Pressure, and Total Static Pressure, the latter being the newer term, as normally understood. The method and apparatus described here, however, does, in fact, take the dynamic side of the equation into account throughout the system as a whole, from main runs to terminal runs as will be described in great detail in the following sections, as this is a key basis of its operation in whole and part.
Catalogued fan systems typically present tabulated or plotted fan data as Total Static Pressure for all intents and purposes and, as a result, the velocity factor is considered secondary, usually assumed as a safety factor. Though a keen design engineer may be aware of this and account for it in the equipment selection and specifications, it is the basis of the following description to emphasize the significance of this velocity factor or “gradient” as it pertains to system operation, after a system is installed and is purported to be under some degree of automated control under normal operation, after the fact.
The Packaged Unit's Total External Pressure
The packaged system's External Static Pressure is, again, a differential of static pressure at the primary system's most exterior intake (before pre-filter section) to its most external discharge side. The purpose of this is to establish the surmountable losses of all internal components within the packaged system, blower itself aside. In basic terms, this measurement is taken from end to end of a packaged unit. Note FIG. 3
Many manufacturers apply this figure instead of what is normally understood as the “Total Static Pressure” of the blower or primary mover. This may be a source of confusion as well, though it may arguably be considered a better starting point in selecting equipment, since it already includes the packaged air handler's own internal losses, which the primary mover must overcome before dealing with any system ductwork/piping/vessel to which it will be connected. For convenience, the engineer, then, need not include additional losses for the internal housing of these systems, though should again be aware of mover characteristics being the heart of a system and the dynamic aspect of this problem, both internally and externally.
The Static Pressure Profile
Beginning from the negative (suction) side intake, a profile is produced with a static, single-point measurement of each key section of the system, sequentially following the path of airflow through to its final discharge into the supply air plenum/duct. FIG. 3 delineates locations for each static pressure sensing point, though these single point or averaged readings, when possible, are taken laterally against the housing wall.
The purpose of this is to obtain pressure drops across each defined section within the packaged system to determine any effectual changes therein as a more detailed analysis. For example, a filter section's pressure drop will rise considerably after it is “loaded” or saturated with dirt and particulate matter. A wet coil will produce a higher pressure-drop than a dry one. These, among other things, will affect total system performance, as well as provide key indicators as to the cause of specific deficiencies and where they originate from within the system. They may point out, for example, the need for a filter change or coil fin cleaning. The type and condition of internal components also affect the primary mover with regard to its ability to deal with any changes occurring external to itself over time and under differing load conditions of cooling, heating, modulating damper control in the mixing box, or other unforeseeable obstructions placed there. Conversely, pressure loss (leakage or undue flow) may be noted there as well.
Normal Mode Vs Smoke Mode Operation
A common oversight in system design involves improperly sizing or equipping a primary mover for all ranges of motion that a mixing box, face-bypass, or other damper control system internal to the unit housing undergoes. This range of motion alters the pressure profile and may place more or less system curve load onto the primary mover. One example: If a primary/secondary air handling system is equipped with both normal mode and smoke mode operation, it will normally produce mixed air (returning and outdoor air combined) at its mixing box to be injected into the building, primary air being the outdoor air portion as building codes and occupancy would dictate. Under smoke mode operation, however, the return air damper closes to 0% and the system will inject 100% fresh air (primary air) into the building to purge smoke, and to work in cooperation with a smoke evacuation fan or other such system in smoke removal. As shown in the following figures, when the path, amount, and temperature/density of entering air shifts from one route to another on the suction side of the unit, the system undergoes a drastic change. FIG. 4 shows normal mode operation within a mixing box, and FIG. 4A shows what typical changes occur in smoke mode operation.
Total Power Available and Required
The key problem arising in the above example is caused by the shift from one duct system to another, each of which has a completely different system curve assigned to it on the suction side and, thus, as a whole system. Adding to this, this is the side where special dynamic losses, known as System Effect losses, most impact the performance of the primary mover in an adverse way. Unlike most losses, these system effect losses associated with dynamic flow occur in such a way that they are not recoverable at any point in the system. They also distort the true performance of the mover and/or system curve. It should be noted that these unique losses cannot be identified by field measurement, only by visual inspection from an experienced Testing and Balancing or Engineering Supervisor.
To begin with, the primary mover and packaged system must be sized bearing the above stated facts in mind, then must be adapted to operate within the framework of changing system conditions. For example, adjustment to minimum conditions should never allow full damper closure due to the necessity of maintaining minimum outside air requirements and free flow (one way or another) that also prevents the suction side ductwork from collapsing, if conversion to 100% suction static pressure or close to it should occur. Ultimately, the correct and final sizing of the primary mover is normally based on the following conditions: lowest minimum outdoor air setting and proportionally minimum return air setting to maintain fresh air and re-circulated air requirements as design and code would dictate. Normally, return air is a fixed setting in its maximum position. Since the advent of single blower systems for supply and return in a single unit housing, most ducted returns fall short of design rates before they would ever increase and, thus, seldom necessitate throttling. This will be further explained in ductwork and fitting losses. Here, the term minimum return air setting provides the most restrictive scenario that a mover might have to contend with, though any additional losses imposed, especially on the suction side of a system should be avoided if not absolutely necessary, again referring to System Effect losses. This could also greatly impact the sizing of the primary mover for little or no reason, further complicated by the effect loss.
Once all total system changes and the normal operating state is clearly determined, the above settings, then, establish the total system curve. This includes all fitted ductwork to and from an established critical run—main and terminal branches intact—needed to be supplied, delivered, and returned by the primary mover to operate at design flow rates, totally and terminally, under maximum demand conditions. Where a variable system is concerned, minimum rates manifest themselves in the form of a system diversity factor, which is further noted.
First and foremost, establishing this initial operating point can prevent the largest and least solvable problem in the initial makings of an entire air or fluid distribution system: over-sizing or under-sizing of total system power required from a primary mover.
Primary Air/Secondary Air Variations
It should be noted that some systems operate only as secondary systems (100% RA, Re-circulated Air or Return Air,) while other systems supply only 100% OA (Outdoor Air,) these being primary systems. Most commercial systems use a mixing box to establish the right mixture of both in one packaged unit, rather than designate another dedicated system to one or the other purpose. Outdoor air requirements are currently 20 CFM per occupant in commercial buildings. Keeping outdoor air to its minimum requirement is generally desirable in seasonal cooling systems, because more outdoor air means more humidity entering the building and more load on the system, thus higher energy demands. Conversely, more re-circulated air means more energy recovered and less load on the air handling unit or any heat exchange terminal. Newer systems employ a mixing box fitted with actuated dampers and sensors which monitor and regulate the entering OA amount when unacceptably high levels of CO2 are sensed in the returning air, this being produced primarily by the exhaling inhabitants of the building. This and other types of controls present a similar problem to smoke mode operation where the system curve and total impact on the primary mover is concerned. These automated systems also directly affect the amount of re-circulated air and cause constantly fluctuating conditions, especially in a VAV (Variable Air Volume) system already plagued with this problem. A modulating OA damper has a minimum setting, never fully closed unless the mode is unoccupied or “off-season,” as some systems would have it. This setting reflects the code requirement for occupancy, and the maximum setting (full open or a specified design maximum rate) is the position taken when high levels of CO2 are detected. The OA setting may be the minimum required or more, not less. As stated before, the major drawback is that more OA=more energy load on the system, unless the example is a heating system operating on an economizer cycle, which takes advantage of cooler outdoor air in such climates. The opposite would then be true, though it is known that hot water systems can maintain as high as 90% of their heat exchange at 50% of hot water flow. The same is not true of cooling systems, which always require at least 80% of their (chilled) water flow to maintain adequate heat exchange.
Consequently, the total RA lowers as the OA goes up. The key terms here are SA (Supply Air,) RA (Return Air,) OA (Outdoor Air.) SA or the total capacity (CFM) of the system is made up of the two components: RA+OA=SA. Also, SA−OA=RA, in this case. Therefore, as one goes up, the other goes down, less total losses or plus gains to the system whole caused by damper positioning changes, leakage, or other internal losses, such as bypassing or infiltration within the unit housing, particularly those equipped with over-sized exhaust fans and relief dampers. The above combined or deducted air equation also applies to older twin blower systems (serving RA and SA independently) when ducted inside the same system, without an exhaust (relief system.) Otherwise, this equation becomes OA=SA−RA+EA when there is an integrated exhaust system.
The Shop Drawing Stage
After a project is approved and building has commenced, the HVAC drawing is usually turned over to a sheet metal fabricator contracted to install the ductwork as true as possible to the engineer's intended design and, later in the process, a certified Testing and Balancing firm is contracted to ascertain this fact, among others, by balancing flow rates within acceptable tolerances, usually 5-10% plus or minus flow rates at terminal outlets and total rates at primary, secondary, tertiary, etc., movers at specified loads with minimal losses.
At this shop stage, a shop drawing is usually produced. This is additional or follow-up drafting work performed by the sheet metal fabricator/installer per “as-built” conditions. It is at this stage, however, that many deviations occur, mainly due to architectural and logistical changes that were never coordinated/scheduled with the rest of the trades on the building project.
This being the case, many fittings, branches, sub-branches are added, taken away, refitted, or entirely omitted as a result. One typical example might be caused by electrical conduits that were run prior to the ductwork being installed and somehow took a wrong turn around where a light fixture was not supposed to be and, hence, blocked the path of an air duct, causing two unplanned elbow fittings to be added where there was supposed to be straight length of run.
Or, it may simply be that an architect decided that an exhaust outlet louver was not aesthetically pleasing on the observable exterior wall of a five star hotel, and so additional length and two 90 degree bend fittings were added to avoid this faux paux. Whatever the situation, these can be taken as typical occurrences on every building project with rare exception.
The ultimate effect of these “as-built” revisions results in system curves changing, sometimes dramatically. And this is the source of most problems on most projects, aside from poorly designed or improperly installed, leaky systems to begin with.
The described method and apparatus may not only assist with this problem, but will become a valuable tool for the system designer and installer throughout the entire commissioning process.
Over all, the best way to counter these recurring problems is for late revisions to be made every step of the way and the described method and apparatus can be involved as early as the computer drafting stage with appropriate recalculations and adjustments pre-programmed to the primary mover and terminal device control panel's memory as they are implemented. Additionally, this process can draw from an entire tabulated database of known equipment, fitting, and performance data as is detailed in this specification. The design operating point will then adjust accordingly against the known flow-pressure constants of the aptly sized primary mover and terminal device(s.)
Key Terminology
Two key types of devices will be discussed: active devices and passive devices. Any motor or otherwise kinetically powered, rotating, pulsating, vibrating, flagellating mover (pump, blower, rotor, etc.) will be referred to as an active device, a device producing force and/or kinetic movement. Terminal, in-line, or discharge devices (variable air volume boxes, valves, monitor stations, diffusers, infusers, registers, grilles, etc.) will be referred to as passive devices. The purpose here is to distinguish between TP, SP, or Vp as actively generated by a mover, or as passively received in an air-fluid stream supplied by that mover.
In air distribution systems, total pressure and its relationship to dynamic losses are expressed as TP(loss)=C×Vp. Total Pressure Loss Equals Coefficient×Velocity Pressure, the coefficient being a tabulation of known fitting losses, such as those provided by ASHRAE publications. Piping head loss in hydronics is expressed as H=FLv SQ./2 gD.
In hydronics, a Cv (valve flow coefficient) is commonly used for valves, terminal devices, and other fittings; while in air systems, a K factor or Ak factor (including free area) is used for grilles, coil face areas, and other terminal flow devices. The above factors indicate losses as they specifically pertain to dynamic flow in either medium and will be referred to as necessary; this to distinguish from provided catalogued data that would only indicate static pressure drops in inches of water column (or gauge) units and the one-sided myopia this may incur.
With regard to Cv's in hydronics, these represent a flow coefficient of a valve or terminal/in-line device in its 100% open position with one PSI of pressure drop across the valve or device itself for standard water, noting that GPM units require no temp./density correction: Cv=GPM/SQ.RT. of Dp (pressure drop must be in PSI units); also, Dp=(GPM/Cv)SQ.; GPM=Cv×SQ.RT. Dp/d (density correction.) Cv's may be established for any hydronics device to be used as a flow meter in so far as catalogued pressure drop data can be relied upon.
K or Ak Factors
Catalogued pressure drops, however, are more in current use in place of K factors where RGD's (Registers, Grilles, Diffusers) are concerned and perhaps for the better. RGD's are the ultimate terminal devices that deliver air-fluid to a given conditioned space. Re-circulated air aside, they are the air/gas/fluid's final destination as far as delivery is concerned. Pressure drops themselves are perhaps a more convenient idea from a design perspective and what it need be concerned with, since K factors are now established under field testing conditions, usually by a Testing and Balancing agency. Terminal devices, however, are inherently dynamic (velocity-oriented) vehicles of air-fluid delivery and should be viewed as such from any standpoint. Due to long time vagaries associated with their proper use, however, K factors are seldom seen in catalogued equipment submittals.
To differentiate the two, a K factor alone is a coefficient associated with a given air terminal device, while an Ak, as noted, includes the free area (cross-section) of that device, factored therewith. At times, these two are used interchangeably, and mistakenly so. This flow coefficient deals specifically with dynamic losses expressed as a diminished free flow area. The K factor simply whittles down the free area to a number less than 1 (a perfect square foot of free flow area) for 12×12 RGD's, keeping in mind that free area is already less than one for those smaller than 12×12. (12×12=144/144=1 sq ft.)
For example, a 12×12 grille (free area of 1) with a K factor of 0.70 (or 70%) has an Ak of 0.70×1=0.70. The Ak includes the free area and may be a number greater than one with larger RGD's and, hence, larger free areas. For example a 12×24 RGD has a free area of 12×24/144=2. If its K factor were determined to be 0.65, then its Ak would be 2×0.65=1.30. This applies to terminal outlets greater than 12×12 or equivalent RGD's.
The K factor is determined by measurement at a terminal flow outlet/inlet with the key equation Q=V×A. Flow equals velocity times area. When a “free” flow rate, albeit in a ducted system, is determined upstream of a terminal or in-line device, along with a face velocity at the outlet discharge of a terminal device, A (or Ak) may be solved for: A=Q/V. If not a free area cross-section, A represents Ak (A & k together) when solved. The K factor alone is not independent of this. If it need be known aside from the free area connected with it, it must be solved separately. The known free area is derived from the nominal dimensions of the cross-sectional duct holding the device without its terminal face RGD, which itself reduces the free area. The K may be solved for alone, or simply put: K=Ak/A
Supply Air Vs. Return Air Distribution
In the case of an exhausting or returning air system, the inlet intake (as opposed to outlet discharge) of a terminal device has differing characteristics. The flow rate upstream of the terminal/in-line device would in this case be on the opposite side, for example, air entering from a conditioned space. This is where free flow rate exists in the form of 100% velocity before encountering the dynamic loss of the RGD.
Velocity readings may then need to be obtained from a traverse of the duct downstream of the grill, moving back toward the primary mover. The flow rate on the face of an RGD is sometimes taken by a barometer (flow hood) reading covering the inlet. Though more questionable in discharge air readings due to taking an air measurement at the face of an RGD after the air stream has already experienced its dynamic losses, this method is widely used by balancers to determine K factors for terminal outlets or inlets out of practical field considerations. Then, of course, Ak=Q (balometer or CFM reading)/V (velocity FPM at RGD face in direction of flow.) Though static and total pressures may have a negative value in exhaust systems relative to atmosphere, velocity pressures or units of velocity, such as FPM, are always thought of as positive values. They are taken in a closed loop differential, High and Low on a micro-manometer facing the direction of flow.
The disadvantage of this distinctly different path of flow and the reason most ducted return air systems fall short of their required flow rates is that they don't have the benefit of ducted total power, and namely static pressure behind them (or rather in front of them) prior to experiencing dynamic losses at the face of their inlets. Leakage rates are also more pronounced on the RA, or EA suction side, where the Vmax (velocity max) is inverted rather than protruded. This also distorts the actual total fan power being applied effectively, as the leaked air still returns to the mover. These, then, are the key differences between the two terminal types and bring to light a problem in current systems with single blower return/supply air. Not to imply that it is impossible to achieve acceptable tolerances, it simply means much less room for error in sizing and fitting return air ductwork and in selecting a primary mover for minimum SA/OA requirements without compromising the RA.
In the case of open plenum (non-ducted) returns, there is less overall restriction, or more dynamic flow at the expense of high, if not complete, pressure loss. Also, there is the distinct disadvantage that return air distribution cannot be precisely controlled, and this is important because it is desirable to return air exactly from zones from where it was distributed in equal measure, less any outdoor air, for optimal recovery. Open systems also suffer from much dirt and outdoor air infiltration from many sources external to the conditioned zones, namely from the equipment room in close proximity to the blower and its open intake. Alternatively, direct-ducted RA/OA systems work best for those that have a smoke control sequence, because less indoor air and, hence, smoke contained therein, may be infiltrated through to the equipment room and re-circulated, despite the best efforts of sealing doors, ceiling plenums, and other adjacent spaces. Partial ducting, a common problem, as with transfer ducts, does not improve the situation and cannot work effectively without direct-ducted fan power—a common oversight in system design. Static pressure is not regained after it is lost through broken duct sections and, at best, this provides only a suggestive pattern of functional return flow through leaky ceiling plenums. Typically, open return systems are susceptible to load mixing from “crossover” zones, discussed later.
Once the true cross-sectional area of a terminal flow device is determined, a non-dimensional velocity passing it (FPM—ft./min., or FPS—ft/sec. in hydronics) is factored to produce a CFM rate of flow (Cubic ft./min.,) or a GPM (gal./min) rate of flow for hydronics, this after the FPS is converted to dimensional cubic ft./sec. units and a minute time frame is applied. This may be expressed as: Q=GPM/60×7.49 (gal/cu. ft. of standard water); also, V (FPS)=Q (cu. ft./sec)/A (cross-sectional area of pipe size.) And finally, GPM=FPS×A×60×7.49.
Piping sizes for fluid flow use the FPS unit, while air systems and standard instrumentation for their testing use FPM units. These are found in traditional tables and charts, which plot head loss against piping length, size, flow rate (GPM,) and velocity (FPS) for various types, such as steel, copper, or plastic pipe. Similarly, air duct tables plot friction loss [“WC (inches water column,) or “WG (inches water gauge) static units] per 100 ft against FPM velocity, flow rate (CFM,) and size of equivalent round duct, this tabulated from rectangular sizes as these cannot be used directly. Noting for emphasis, both types of charts are plotted against friction loss only (a static unit of measurement,) as it would relate to length of run, or equivalent length of run, this to isolate the dynamic aspect of system sizing and design which has to do with fitting/directional losses and reduced area coefficients. This is the industry standard terminology using the inch/pound system, which will be the choice of this specification, though the described method and apparatus may also function in metric equivalent units, if desired.
Among other pitfalls of designing and maintaining an air-fluid distribution system, the problem with catalogued K factors and any other such air-fluid flow coefficients, is that the data may be largely erroneous due to misrepresentation of actual field conditions, the point being that the K factor is unique to a given system and must be established by field testing of that system, as opposed to tests conducted under “ideal,” static lab conditions. This is particularly true of plenum box or soffit-type vessels with sidewall registers or grilles connected perpendicular to airflow and connections generally not in the direction of flow. Many of these infinite dimensional variations would never or could never be reproduced under lab conditions. In fact, there are simply too many possibilities and variables within a system to warrant such constancy, as it can never be possible, especially with the unpredictable nature of “as-built” conditions caused by late shop changes to ductwork, capped extensions, turbulence or non-laminar flow, and other un-contoured paths of air-fluid flow.
Another issue with K factors involves their use in VAV systems in adjusting the sensed flow versus actual flow to a terminal branch via a terminal branch device (VAV box, zone damper, valve, etc.) Currently, most leading systems are equipped with adjustment of a K factor or K “value” for given terminal branch flow characteristics. This may be adjusted by a Balancer to calibrate the terminal device's sensor to what flow is actually not only passing the control device/flow monitor station, but reaching each terminal outlet, the final destination of delivery. The difference of these two, sensed versus actual, indicates losses due to leakage, dynamic losses, or friction losses—one of these three. Normally, the balancer has only to enter the sub-total flow reading he ascertains per outlets for that branch with his own timely calibrated equipment and enter this data into the control system, which makes the basic adjustment: Actual flow/Sensed Flow=K value used to adjust sensor reading and, thus, damper position.
If this value is less than 1, then the flow rate is less than the sensor indicates. If this value is greater than one, flow is more than sensor indicates. The sensor is then calibrated based on this entered data reflecting actual system conditions by calculating a new flow coefficient that reflects unique system losses for that particular branch. However simple this process may seem, it still belies the fact that the system must work harder, terminally and totally, to achieve the flow rates due to system losses producing flow factors that may be unacceptably low. Typically, these may fall between 0.65 and 0.80 and rarely, if ever, produce factors at or above 1.
Prior to the balancing procedure, the controls contractor or supplier presets the terminal device with a factory setting per design specifications at the outset of the project. In current practice, the terminal device is roughly sized for a flow capacity-range, or at least as closely as stock sizing will avail. Afterwards, the device seeks to establish this setting with it own sensing faculties and maintain what it believes to be the correct setting until it is told otherwise by a user.
The above procedure establishes the main user-control system interface where those skilled in the art are primarily concerned, though a control contractor may be more attentive to zone temperature settings and changes, and, above all, achievement of those settings one way or another, whereas a Testing and Balancing contractor is concerned primarily with air-fluid flow rates, in both total capacity and terminal capacity.
Noted discrepancies between design capacity and actual performance, however, are due to the system characteristics of the ductwork/piping/vessel downstream of that terminal device not readily apparent due to current control sensing limitations. In some cases, improperly placed, connected, or malfunctioning sensors could also distort actual conditions. The former may stem from late changes made to the terminal branch, unexpected losses due to obstructions, acute bends or turns, changes to sizing of the terminal device for its range and capacity versus any revised terminal branch system requirements, etc. Additionally, an effect caused by downstream throttling of terminal or takeoff branches contributes to adverse effects, as this may confuse current flow sensors, which, contrary to popular belief, are more precise in taking measurements in closer proximity to the terminal/in-line device or flow station at which they are situated.
What Goes in does not Come Out
Consequently, where flow-volume is concerned, “what goes in does not come out,” contrary to widely held belief. This goes for system total or terminal branch. The difference results from losses in one of three forms: leakage, friction losses (SP), or dynamic losses (Vp.) Perhaps the denial exists due to the fact that the primary mover is a “constant volume machine” as long as rotation is constant. However, aside from leakage, nothing is truly lost, but rather converted. Curve riding and changes to a mover (namely speed of rotation) versus changes to a system (length or fitting) also explain this phenomenon. This also stresses the importance of why these relationships must be viewed in the context of an operating curve and not independently, as they tend to be.
The key problem, however, lies in the issue of making best use of this conversion. Much of this has to do with the improper pairing of a mover with its system, or a terminal device with its sub-system, and the claims address this problem as supported by this description. Most commonly, the losses are a result of leakage, but when the expected volume “does not come out,” the remainder may be deemed as static pressure resulting from undue restriction. Essentially, potential energy pent up inside the system is not yet or perhaps never released as flow. It does, however, exist dormant within the system so long as mover power is applied. The applied force will also exist as long as the ductwork can contain it for its class and rating. Otherwise, it becomes leakage at one or more points in the system.
One adverse result of this is that more input power must be applied to achieve the same flow rates at terminal outlets. When applied deliberately, however, static pressure may be manipulated to produce intended results, as is discussed in embodiments. Main and terminal branch problems are also further examined in the section on “Upstream Leverage,” an additional supporting claim on the said method and apparatus, and in the section on terminal device flow control and all problems associated with this.
Overall, the issue of K factors, Cv's, or flow coefficients in general is an additional supporting concept for the said method and apparatus, referring in particular to terminal devices and their characteristics within a given, real system, as opposed to a theoretical one. Lab testing and equipment cataloguing also stand to benefit from implementing this method and apparatus at the very outset.
Current Use of ATC: DDC-AD Conversion
Among previously mentioned problems, current DDC (Direct Digital Controls) also suffer from quite severe limitations imposed by their very linear nature, namely the linear nature of the micro controllers they are comprised of, because mechanical, thermal, and fluid dynamic relationships are anything but linear. This points out another key advantage of the described method and apparatus: complex curves and relationships are plotted first and foremost, then coordinated data is processed after this crucial process and other key processing occurs.
Affinity laws alone do not apply to movers outside of a controlled context, only theoretically speaking, where direct, squared, and cubed relationships are concerned. And when they are, they rely heavily upon extrapolation, rather than interpolation. However, where actual field-testing is concerned, these conditions always vary and stray quite abroad, especially at low and high ends of the spectrum when dealing with a lab-tested mover in the constantly changing framework of a real, “as-built” system.
In the proposed system, heat flow is plotted using psychrometric principles, namely tabulated data in tenths of degrees. Affinity relationships governing the mover will be displayed on graphs and are used to plot actual performance curves, as opposed to how they might perform theoretically at varying positions of WOAF (Wide Open Air Flow.) FIG. 6 and FIG. 6A.
Following this initial pairing of system to mover, true coordinates are determined, then translated into readable data as required by a logic-oriented micro-controller. This point also conflicts with current use of temperature sensor-oriented controls, which are not governed by the affinity laws or even thermal dynamics. They simply operate on the direct linear scale of the micro controller, using single integer math, or operate some form of motor control to effect conditioning changes, normally on a proportional (direct-acting) interface between motor controlled damper-actuator and basic sensors. The key problem remains, however, that they go little or no further in obeying the laws of thermal dynamics or fluid mechanics, or in making use of them for efficiency or effectiveness.
As shown in FIG. 10, the described method and apparatus uses plotted coordinates established with known affinity laws as a starting point and guided by them whenever unknowns are present. This can then offer a complete picture where there may be missing links or data unavailable. Following this, the transfer of data inputs and outputs can then be adjusted correctly to perform the necessary functions as required by the hardware. However, this description emphasizes that in using the described method and apparatus, no unknowns will cause an extrapolation to become necessary. Between the breakdown of Total Power and Total Pressure, there shall always be a solid deduction (as opposed to induction) made never contingent upon unknowns.
Most industrial sensors still require AD (Analog to Digital conversion,) and so are technically not “directly digital,” as the name would suggest. Such sensors still require transduction at some point to convert an inherently analog signal, for lack of a better term, to a code palatable to a microprocessor. The crux of the problem lies in correct sensor interpretation and signal utilization. Characteristic and performance curve plotting based on proper sensor placement, input, and configuration is the best approach. This may be done first by true sensor feedback based on correct thermal and fluid mechanics principles, curve plotting, then processing, as explained with said method and apparatus in this specification. Any other method, therefore, must be assumed to be grossly limited, if not wholly incorrect, particularly if based on principles of temperature zone sensing and direct damper control alone with localized, unilateral feedback.
In summary, the prevailing difference between the described method and apparatus and current systems lies in temperature control with direct digital motor control alone versus complete fluidic control; thermally, statically, dynamically, and totally.
Key Prime Mover Types and Configurations
Generally, there are two types of movers at either end of a wide spectrum: High-pressure type and Low-pressure type. An archetypal example of a Low-pressure type air mover would be the basic propeller fan or axial fan. Typically, this moves air at a high velocity, high volume (CFM) and does so at the expense of static pressure. Vane Axial or Tube Axial may be easily confused with Radial in-line fans, which are actually centrifugal and sometimes referred to as the same or may appear similar, though they are not. A radial fan's blades don't stem from the shaft, as with a vane or “prop,” but a radial ring of blades rotates about the interior housing rim. They are however, SWSI (Single Width, Single Inlet) and in-line with the ducting much like Vane Axials. The most typical example is the outlet-capped, “mushroom” fan that generates high end-suction typically used in rooftop exhausts.
On the opposite side of the spectrum, the centrifugal fan and its variants produce higher static pressures with less flow-volume output, comparatively speaking. The FC (Forward Curved) and BI (Backward Inclined) fans are two key types of centrifugal fans, each with desirable and undesirable characteristics of their own. BI type fans are an example of a higher-pressure type blower, while FC's, used most commonly for commercial applications, are a compromise of pressure and flow (or velocity content, which translates to flow.) Most centrifugals are DWDI (Double Width Double Inlet) for maximum flow-through capacity and air movement volume at given pressures, though even higher-pressure types are narrow, single-inlet designs for dust, particle collection, or other high suction vacuum applications. Again, with loss of flow-volume under applied motor force, there is pressure gain, whether suction or discharge. There is also more demand on brake horsepower with this configuration.
Whatever the traits of each type of mover are, its general performance characteristics are displayed on a “characteristic curve” and each is suited to a specific application. In current usage, this identifies specific qualities and desirable operating points for flow-volume rates at given static pressures and maximum “static efficiency,” which is a concept that is flawed from the inception of equipment cataloguing, along with percentage of WOAF, also a static, theoretical projection of mover-system performance that completely misuses the dynamic gradient. Percentage of closure testing as currently in use has known, acknowledged failures and in no way substitutes for real system characteristics and/or how the mover reacts to those unique characteristics in actual field operation. As currently accepted, most FC fans' operating ranges fall on their 60% of wide open flow for peak static efficiency, still providing adequate flow rates, while BI fans have a non-overloading (amperage) characteristic and a higher static efficiency at the expense of lower flow rates. In terms of their pressure content, the FC fan produces approximately 20% SP (Static Pressure) and 80% Vp (Velocity Pressure,) while the BI fan produces approximately 70% SP and 30% Vp. This theme of specific flow-pressure content will be referred to throughout this specification. FIG. 5 shows typical performance curves for various fans.
The described technology proposes an integrated fluid control unit and metering device equipped with self-calibration through all system load variation as required by changing scalar or vector flow coefficients, including Brake Horsepower, critical Total Pressure, and Critical Mass Flow as consummately applied.
In support of this current novelty, many factors place prior art in question. One popular misconception in flow testing and mover control is that the mover's RPM will change as dampering differences or relief openings are imposed on a distribution system. For example, one may feel that if they open an access panel with the blower running—and release Static Pressure—that, along with a notable increase in amperage, the mover's rotation will also increase. This is not so. The mover speed of rotation and unique loading characteristic is independent of the system (unless it is changed in of itself) and it is precisely for this reason among others that the relationship must be viewed in a context that properly adjusts these changing parameters, further including BHP or Total KW.
Basically put, changes to one conform to the other in a curve-riding relationship along corrected sine/cosine tangents/cotangents. This offers a comprehensive way to control and monitor such a fluid handling system and expect to achieve predictable results. This may also be expressed through PHI, phase angle on the electrical side, clocking signal under modulation, or effective damper angle for a valve or terminal device under modulation.
Variable geometry also figures in converging or diverging angle fittings for fixed ducting or opposed blade dampers. Otherwise, changing valve coefficients (10) are precisely tracked and pinpointed by degree opening or effective radian angle (5) as shown on the quadrant chart example (FIG. 11) for the terminal device and its constant (11). In electrical signal modulation, this chart simply spans 360 degrees and two or more Operating Points are in play, such as with total system parameters (23, 24) for a moving signal or waveform.
In prior use, certain physical laws known as affinity relationships were employed to estimate the performance of such fluid systems through an extrapolation (educated guess) as to how the actual system may perform under given conditions (FIG. 10). These, however, were simply projections based on presumptive logic and guesswork. The described method takes appropriate measures using interpolated data, deducting the solution from three or more known and firmly established verification points.
By virtue of pure logic, one novelty of the described technology is that it need never rely on any extrapolation (educated guess) to determine true performance characteristics. The procedure will always conform a precise deduction from BHP or Total KW calculating steps, as these parallel Total Pressure and its subsequent conversion into Velocity Pressure (Vp) and Static Pressure (SP). This offers the basis of a new form of logic gate for fluid-mechanical systems. It also proposes a computer operating system for virtual and real physical environments where in place of the “cursor”, a point or points of operation are interpolated by the processor for the appropriate physical actions, whether scalar or vector in nature.
In current systems, so-called “floating” data points tend to be viewed independently and compound errors result. Current systems utilize extrapolative performance projections based namely on Static Pressure sensing with sensors also placed in a questionable context, both up and downstream of dampering or other variables where correct interpretation is rendered inaccurate and unreliable. Movers and valves can only “hunt” for an obscure range or point of operation from conflicting sensor data as pressure increases can be as equally attributed to block-tight Static Pressure as they can be to fan power being applied effectively. This also easily confuses the blower because most typical centrifugal fans exhibit the same Static Pressure characteristics despite a vastly different flow rate, at approximately their 30% and 70% points of “Wide Open Flow”, known as their surge points. This is especially pronounced on the low and high end of the curve where the motor's Power Factor is also not made use of appropriately. This problem explains “blower surge”, however, the method algorithm also addresses the phenomenon known as “system surge”, another adversity in fluid systems.
Though the described Operating Point may be placed in any desired field for efficiency or effectiveness, its prime function also accounts for “Fan Horsepower”, “Air Horsepower”, and “Water Horsepower”, additional forms of BHP denomination, as well as overall “Mechanical Efficiency” where the unit “driver” and “driven” components are in play. This covers any internal drive losses as well as polytropic effects imposed by the compressible or incompressible state of fluids.
Efficiency is usually the biggest questions mark in such systems, because it is often obtained from a manufacturer's said tag HP (not BHP) or some previous estimation. Mechanically, this component may also be derived from sensor data where BHP is first determined by alternate means such as on a torque gauge along with RPM readings; Torque (lb-ft)×RPM/5252. Mechanical output, however, is appropriately determined and distributed via the sensing apparatus from Total Pressure conversion as produced by system load under specific variation. ME (Mechanical Efficiency)=AHP (Air Horsepower/BHP; or WHP (Water Horsepower)/BHP; any fluid stream power/BHP.
Electrically, a direct Power Factor reading (KW/KVA) or P/S can be taken and remaining electrical unknowns are derived from the power triangle consisting of P, S, and Q (True Power, Apparent Power, and Reactive Power, respectively). The Pythagorean Theorem follows in this relationship where Q (reactive)=SQ. RT. S SQ.−P SQ. and so forth. Additionally, comparative data may be derived from Mechanical Efficiency to assess the electrical-mechanical translation of these components.
Power Factor is central in assessing electrical power output, along with electrical efficiency—power available for useful work, as opposed to KW input. But between power draw from the mover and translations of Total Pressure, the actual unit efficiency is accurately determined in a real system as opposed to a “proposed” efficiency, whether mechanical or electrical. Also, BHP may be derived from input KW (voltage and amperage readings) where only the Power Factor is known, this determined by direct Power Factor reading, input KW/KVA, or other means. KW output=IXEXPF/1000 (single phase power); or IXEXPFX1.732/1000 (three phase power). Once true power output is assessed, then electrical Efficiency=746XBHP/EXIXPF (single phase power); or 746XBHPXEXIXPFX1.732 (three phase power). If this were “proposed” efficiency, then BHP would be tag or manufacturer “HP” and estimated “PF”.
Velocity reading as per pitot tube multi-point traverse is deemed among the most accurate datum points with its closed-loop sensing, second to BHP. Static reading is deemed the least accurate. Additionally, Static Pressures are prone to atmospheric differences inside of a building envelope (highly significant at 14.747 PSI) when used out of context of these other crucial data verification points. This discrepancy in itself can equal the addition or absence of a large capacity mover. This unacceptable margin for error can easily be breached if such pressures are not viewed as “absolutes”, taking an atmospheric reference into account at both manufacturing stages and at final testing stages of an “as-built” system.
Under VAV operation, the method algorithm performed by the said apparatus establishes a set criteria for the “System Diversity” amount—the specific energy saved—and the control system may itself “map out” this diversity through its own default operation setting as most effective for an existing or “unknown” system. Solved unknowns are extracted from precisely coordinated relationships using the said verification data points. The diversity manifests itself in minimum requirements for all loading demands and minimum valve positioning in a real system.
The Diversity is a valuable amount of the distribution system that can be set aside when not in use, a margin for saving energy, when portions of the mover and system are not in full demand instantaneously or, in other words, “not instant.” Current methods of “instant” reading or sampling flow and pressure data, however, cannot keep up with these complex changes, namely due to a problem known as “flow-pressure stability” and other analog-digital control limitations. These can be viewed on a power triangle signal graph. Logging these clocked leading and lagging “trends”, this adverse effect becomes increasingly apparent on the fluid control side of the equation and then reverberates through a cascading effect through all high and low voltage electrical systems, including microprocessors as well. The described technology offers a solution to this inherent problem on a fluid-mechanical, thermal, and electrical level.
Because critical areas of a fluid system change under modulation, the mode of operation continually adjusts the total circuit path and its demands on the mover, which fall into play precisely where needed at any given time or constant as the ordinate, abscissa, and “sigma” sensor values would indicate (FIG. 13). This is especially crucial in air systems due to their changing flow coefficients with adverse effects imposed by damper modulation and damper angle adjustment. Due to limitations of current systems, valves operate within only a small part of their usable range. Utilizing the specified method algorithm and prescribed apparatus, the variable mover and plurality of valves are placed in the broadest and most effective range possible within the given system.
Aside from the VAV Mode, other specified modes, notably Test Mode, Balance Mode, and Smoke Mode, simply use similar terminal device or main dampering techniques to effect other actions. Lab Test, then Balance Modes would apply from initial lab testing stages through to start-up, troubleshoot and calibration of the system as needed. “WOAF” (Wide Open Air Flow) originates from the nascent stage, where initial data points are first established and recorded in the database provided, or derived from some other accepted source. Smoke Mode is triggered by a condition in a built-up system of fire smoke evacuation in which all valve variables are at wide open parameters, namely 100% O/A (Outdoor Air) injection, but fully closed R/A (Return Air). As added measures, the remaining functions deal with eliminating leakage and “System Effect” factors through isolated sensing and dampering techniques as specified.
The Expansion-Compression Cycle
The fluid metering and control unit also applies optimal functioning in refrigeration systems where the DX expansion-compression cycle is used. Here, the terminal device or heat exchanger may be a vessel of compression or a vessel of expansion. This subject matter pertains to compressible fluids or gases where a polytropic process is assumed along with air-fluid changes occurring above atmosphere as well as those below, such as in vacuuming (suction) applications. Critical mass flow rate and timing through the heat exchange refrigerant coil, expansion valve, water coil, or other HX medium are also precisely controlled this way through functions pertaining to heat exchange of diverse fluids crossing paths with one another in different configurations, counter-flow being the most effective.
In summary, the path of critical mass flow in variable systems is precisely manipulated and tracked by the “Point of Operation” reference point, expressed as either a scalar function or a vector function. This complex coefficient maintains an adequate flow-volume-pressure relationship in the whole system, totally and terminally, thus satisfying the need for system diversity on a fluid-mechanical and thermal dynamic level.
Moreover, the key utility of this patent provides the means of “tuning” most all machines and mechanical devices for operating at their optimal level of power and efficiency at any given time or constant. This includes fully articulated operation through all varying volumes, densities, variable geometries, and, ultimately, critical mass flow rates at their maximum possible effectiveness.